Where image encoding is implemented by hardware, generally a frame memory for storing a reference frame is frequently incorporated as an external DRAM (Dynamic Random Access Memory) chip separate from an LSI (Large Scale Integration) for encoding. Since it is necessary for such a frame memory as just described to store a plurality of reference frames or to be accessed at a high speed in a process such as motion search (ME (Motion Estimation)), MC (Motion Compensation) or the like, it is necessary for the frame memory to have a sufficiently great data storage capacity and have a sufficient broad frequency bandwidth for inputting and outputting data.
However, due to 4K televisions or increase of the capacity of image data in recent years, there is a tendency that the data amount necessary to be handled by an image encoder increases. Therefore, also an external DRAM chip is demanded to have a greater capacity and a broader frequency bandwidth, and this makes a cause of increase of the product cost.
Therefore, a method for compressing and storing image data has been devised (for example, refer to NPL 1 and NPL 2).